First synthesis of racemic saphenamycin and its enantiomers. Investigation of biological activity

Article abstract of DOI:10.1016/S0968-0896(02)00472-8

The natural antibiotic saphenamycin, 6-[1-(2-hydroxy-6-methyl-benzoyloxy)-ethyl]-phenazine-1-carboxylic acid, was synthesized from saphenic acid using temporary allyl protection of carboxy and phenoxy functionalities. Resolution of racemic saphenic acid was performed by crystallization of the corresponding (-)-brucine diastereomeric salts and the absolute configuration of (-)-brucinium (-)-saphenate was determined by X-ray crystallography to have R-configuration. This also proved to be the configuration of natural saphenic acid. Enantiomers of saphenamycin were obtained from resolved saphenic acid and screened against a range of skin flora and resistant Staphylococcus aureus strains. Biological activities of saphenamycin enantiomers were compared with that of the synthetic racemate as well as earlier reported activities of saphenamycin isolated from natural sources. No significant difference was observed in activity of the enantiomers of saphenamycin, which revealed that the chirality of saphenamycin has no consequences for the antibiotic activity. Saphenamycin proved to be a potent antibiotic against fusidic acid and rifampicin resistant S. aureus strains showing MIC of 0.1-0.2 μg/mL. 2002 Elsevier Science. All rights reserved.

Full text of DOI:10.1016/S0968-0896(02)00472-8

Bioorganic & Medicinal Chemistry 11 (2003) 723–731
First Synthesis of Racemic Saphenamycin and Its Enantiomers.
Investigation of Biological Activityy
Jane B. Laursen, Charlotte G. Jørgensen and John Nielsen*
Department of Chemistry, Technical University of Denmark, Kemitorvet, Building 207, DK-2800 Kgs. Lyngby, Denmark
Received 26 July 2002; accepted 27 September 2002
Abstract—The natural antibiotic saphenamycin, 6-[1-(2-hydroxy-6-methyl-benzoyloxy)-ethyl]-phenazine-1-carboxylic acid, was
synthesized from saphenic acid using temporary allyl protection of carboxy and phenoxy functionalities. Resolution of racemic
saphenic acid was performed by crystallization of the corresponding (À)-brucine diastereomeric salts and the absolute con-
figuration of (À)-brucinium (À)-saphenate was determined by X-ray crystallography to have R-configuration. This also proved to
be the configuration of natural saphenic acid. Enantiomers of saphenamycin were obtained from resolved saphenic acid and
screened against a range of skin flora and resistant Staphylococcus aureus strains. Biological activities of saphenamycin enantiomers
were compared with that of the synthetic racemate as well as earlier reported activities of saphenamycin isolated from natural
sources. No significant difference was observed in activity of the enantiomers of saphenamycin, which revealed that the chirality of
saphenamycin has no consequences for the antibiotic activity. Saphenamycin proved to be a potent antibiotic against fusidic acid
and rifampicin resistant S. aureus strains showing MIC of 0.1–0.2 mg/mL. #2002 Elsevier Science. All rights reserved
# 2002 Elsevier Science Ltd. All rights reserved.
Introduction
activity,6,8 antitrichomonal activity,5 and larvacidal
activity against mosquitoes.5
Bacteria and fungi synthesize a plethora of biologically
active natural compounds with functional characte-
ristics such as antibiotic and antitumor activity. Several
of these compounds are believed to act directly on
nucleic acids modulating replication, transcription or
translation processes.1À4
Saphenamycin contains the flat tricyclic aromatic phen-
azine chromophore, which has been found in several
other secondary metabolites isolated from various bac-
terial genera.9 Many of these are known to be anti-
biotics and virulence factors.9 Recently, we reported the
synthesis of new analogues closely related to saphena-
mycin. One analogue exhibited activity against several
bacteria with potency similar to saphenamycin.10
Among these compounds is the yellow pigment and
secondary metabolite saphenamycin (1), 6-[1-(2-
hydroxy-6-methyl-benzoyloxy)-ethyl]-phenazine-1-car-
boxylic acid (Fig. 1), which was isolated from a strain of
Streptomyces by Michel et al. in the early 1980s and
later also by other research groups.5À7 Saphenamycin
was found to exhibit a broad range of biological activi-
ties including antibacterial activity against Gram-posi-
tive and Gram-negative bacteria,5,6,8 antitumor
The antibiotic action of phenazines and other planar
aromatic molecules may be complex but intercalation
yThis work was presented in part at the 223rd National Meeting of the
American Chemical Society, Orlando, FL, USA, 7–11 April, 2002,
ORGN-243.
*Corresponding author at current address: Department of Chemistry,
The Royal Veterinary and Agricultural University, Thorvaldsensvej
40, DK-1871 Frederiksberg C, Denmark Tel.: +45-3528-2436, fax:
+45-3528-2398; e-mail: jn@kvl.dk
Figure 1. Structure and general atom numbering of saphenamycin (1).
0968-0896/03/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(02)00472-8

724
Jane B. Laursen et al. / Bioorg. Med. Chem. 11 (2003) 723–731
into duplex DNA has been proposed for several phena-
zine antibiotics.11,12 Also, generation of toxic anionic
radicals involved in redox cycling has been suggested as
a mode-of-action for phenazine-based antibiotics.9
hands, preparation of the sterically hindered acid
chlorides, unprotected as well as methyl- and allyl-pro-
tected, proceeded smoothly under ‘Vilsmeier’-conditions
using oxalyl chloride in the presence of a catalytic
amount of DMF.16
Saphenamycin contains one asymmetric carbon atom,
the secondary benzylic position (C10, Fig. 1). Early X-
ray crystallographic studies performed on saphenamy-
cin isolated from natural sources by Kitahara et al.6
revealed only the relative structure of saphenamycin
and specific rotations of isolated samples were close to 0
and very inconsistent.5,6,8 Until now, no reports have
appeared on the absolute configuration of naturally
occurring saphenamycin.
Protection of both saphenic acid (3) and 6-methyl-
salicylic acid (2) seemed crucial for the synthesis in
order to minimize formation of byproducts. Esterifica-
tion of allyl-protected saphenic acid (4) with excess of
unprotected 6-methylsalicylic acid gave a ‘double ester’
5 of saphenamycin in 64% yield and only limited
amounts of the desired partially protected saphenamy-
cin (Scheme 1).y
The present paper reports an efficient synthesis of
saphenamycin as a racemate using allyl protection.
Furthermore, the successful resolution of saphenic acid
and elucidation of its absolute configuration has led us
to the synthesis, identification and biological testing of
the enantiomers of saphenamycin. To the best of our
knowledge this is the first report on elucidation of the
absolute configuration of these naturally occurring
phenazine derivatives.
Synthetic strategy
Simple disconnection of the ester bond in saphenamycin
gives two moieties; 6-methylsalicylic acid (2) and
saphenic acid (3). When appropriately protected these
moieties will couple to form a protected form of saphena-
mycin. Allyl protection proved ideal for accomplishment
of the synthesis.
Scheme 1. Formation of ‘double ester’ byproduct when esterifying
allyl saphenate (4) with excess 6-methylsalicylic acid (2).
Moreover, adding unprotected saphenic acid (3) to o-
toluic acid chloride resulted in several byproducts. Use
of identical protecting groups for phenol and carboxylic
acid reduced the number of deprotection steps. Fur-
thermore, earlier reports had drawn attention to steric
hindrance of the ortho-substituted benzoic acid (vide
supra). This emphasizes use of a sterically less deman-
ding protecting group for the phenol that is essential for
satisfactory reactivity of the acid. Additionally, ortho-
gonality of protecting groups and the benzylic ester in
saphenamycin was essential. Thus, the small and easily
removable allyl group17 was utilized for protection of
hydroxyl as well as carboxyl functionalities in the
synthesis of saphenamycin (1).
Results and Discussion
Synthesis of 6-methylsalicylic acid (2) and racemic
saphenic acid (3)
Several methods for preparing 6-methylsalicylic acid (2)
have been reported. The method described by Hauser et
al. in 198013 provided 6-methylsalicylic acid ethyl ester (6),
the first starting material in our saphenamycin synthesis.
The Hauser method involved a Robinson annulation
reaction of inexpensive starting materials, crotonic acid
and ethyl acetoacetate. Bromination followed by dehy-
drobromination gave 6-methylsalicylic acid ethyl ester
(6) in a few moderate to high yielding steps.
Saphenic acid was protected as the allyl ester (4) in
quantitative yield using a method described for amino
acid derivatives.18 Allyl-protected 6-methylsalicylic acid,
2-allyloxy-6-methylbenzoic acid (8), was obtained from
a Williamson allylation of 6-methylsalicylic acid ethyl
ester (6) followed by hydrolysis of the ethyl ester (7) in
92% yield for the two steps (Scheme 2).
In 1999, we reported an improved synthesis of racemic
saphenic acid (3) using a non-stereoselective reductive
ring closing method.14 This provided us with the second
starting material for racemic saphenamycin synthesis.
O-Allyl-6-methylsalicylic acid (8) was treated with oxa-
lyl chloride and DMF, forming the acid chloride quan-
titatively. The presence of the corresponding acid
chloride was confirmed by NMR, which showed a
change in chemical shift of the 6-methyl groupfrom
2.46 to 2.38 ppm in agreement with reported NMR data
Synthesis of racemic saphenamycin (1)
Early attempts by Bahnmuller et al. to synthesize
saphenamycin from unprotected saphenic acid (3) and
protected as well as unprotected 6-methylsalicylic acid
failed.15 The lack of success could be ascribed to inade-
quate reactivity of the sterically hindered 6-methylsa-
licylic acid derivatives towards thionyl chloride. In our
yLimiting the excess of 2 (from 4 to 2 equiv) led to formation of several
byproducts and even lower yield of the desired product.

Jane B. Laursen et al. / Bioorg. Med. Chem. 11 (2003) 723–731
725
Scheme 2. Synthesis of saphenamycin; (a) All-Br, K2CO3, acetone, reflux 24 h (94%); (b) NaOH (s), EtOH/H2O, reflux, 30 h (98%); (c) All-Br,
DIEA, CH3CN, rt 18 h (quant.); (d) (1) (COCl)2/DMF, CH2Cl2, 2 h; (2) DMAP, pyridine, room temp. 24 h. (64%); (e) Et2NH, Pd0(PPh3)4 (cat.),
dioxane, rt 2.5 h. (88%). 88% of the R-isomer was obtained using optimized conditions as listed in this scheme. The racemic saphenamycin was
prepared using non-optimized conditions that yielded 61%.
of O-methyl-6-methylsalicylic acid and the correspond-
ing acid chloride.19 Furthermore, one dropof the reac-
tion mixture was treated with methanol and the methyl
ester was visualized by TLC analysis. The acid chloride
was concentrated in vacuo and used directly without
further purification.
demonstrated the importance of full conversion in the
deprotection step in order to obtain acceptable yields.
Using 2.25 equiv of a simple secondary amine, diethyl-
amine, and 0.15 equiv of Pd0 full deprotection and up to
88% isolated yield was obtained.
Characterization of compounds
Conditions for the base-catalyzed esterification of
saphenic acid allyl ester (4) and O-allyl-6-methylsalicylic
acid chloride were investigated. Reaction in CH2Cl2
with DMAP as a catalyst and Et3N as stoichiometric
base was sluggish. More polar conditions using DMAP
as a catalyst and pyridine as both solvent and base
afforded the allyl-protected saphenamycin (9) in good
yield (64%).
The phenazine compounds were characterized by MS
and NMR and a full assignment is included here to
resolve contradictions in earlier reported assign-
ments.7,20,21 The relative assignment within the groups
of protons 2–4 and 7–9 could be established by COSY
spectroscopy. The complete assignment of protons on
the phenazine ring was achieved using gHSQC and
gHMBC spectroscopy. Especially, the 3-bond coupling
between the carboxyl-C and H-2 was useful.
Deallylation of compound 9 relied on Pd0 catalysis via
formation of a presumed p-allyl palladium(II) complex.
Regeneration of Pd0 can be achieved by either hydride
donation or cleavage of the allyl palladium species by
addition of another nucleophile. First, Pd(PPh3)4 was
used in catalytic amount as palladium source and the
use of several conventional external hydride donors
(NaBH4, Bu3SnH and PhSiH3) was investigated, but all
reactions failed. This may be ascribed to the lack of free
coordination sites on the palladium source used. Next,
internal hydride donation from formate with triethyl-
amine as base was investigated, and in this case the allyl
ester was readily cleaved, while conversion of the allyl
ether was sluggish. Cleavage of the allyl ester creates the
nucleophilic carboxylate that competes with formate in
the coordination to bis(triphenylphosphine)-palladium
(II). However, it is still likely that the allyl ester is
cleaved without palladium catalysis by direct trans-
esterification to form the allylformate. Saphenate car-
boxylate coordination may hinder coordination of
formate, hence the intramolecular hydride delivery from
formate to PdII and locks the catalyst in its oxidized
state inhibiting the catalytic cycle. Finally, an external
nucleophile was added to successfully trap the PdII-
bound allyl and regenerate Pd0 without use of addi-
tional coordination sites on palladium. In our hands,
separation of saphenamycin (1) from the corresponding
aryl allyl ether proved to be particularly difficult. This
It was observed that esterification of the 10-OH with
protected 6-methyl salicylic acid greatly affected the
shift of H-10 (from 5.7 to 7.5 ppm). Moreover, the H-2
shift changed dramatically (from 8.3 to 9.0 ppm) upon
deprotection of 9 to saphenamycin (1) (Table 1).
The assignment of H-2 and H-4 in allyl saphenate (4) is
in agreement with the assignment of methyl saphenate
made by Van’t Land20 and Yun21 but it differs from the
assignment of these protons in methyl saphenate made
by Keller-Schierlein et al.7
Resolution of saphenic acid (3)
With racemic saphenic acid already at hand,14 effort was
focused on resolution of racemic saphenic acid rather
than on developing a stereoselective synthesis. Initial
experiments with separation of the enantiomers of
saphenamycin (1) and saphenic acid (3) on semi-pre-
parative chiral HPLC provided us with small analytical
samples. The amounts of 1, however, were inadequate
to fully characterize and analyze the enantiomers of the
antibiotic, and the amounts of 3 inadequate to perform
further synthesis in order to obtain the enantiomers of
saphenamycin. Hence, methods for resolution of saphe-
nic acid on larger scale were considered.

726
Jane B. Laursen et al. / Bioorg. Med. Chem. 11 (2003) 723–731
a
Table 1. 1H NMR data (ppm) in CDCl3
Compound
H-2
H-3
H-4
H-7
H-8
H-9
H-10
H-20
H-300
H-400
H-500
H-700
2.55
6
6.84
8.0
6.74
8.5
6.82
8.4
7.26
7.6, 8.2
7.19
8.0, 8.0
7.28
7.5, 8.4
6.71
7.7
6.79
7.5
6.88
7.5
3J(H,H) Hz
7
2.30
2.48
3J(H,H) Hz
8
3J(H,H) Hz
4
8.26
6.9
8.28
6.8
9.02
7.0
7.77–7.88
m
7.85
m
8.08
8.34
8.7
8.46
9.0
8.61
8.8
7.77–7.88
m
8.04
6.5
8.01
m
7.77–7.88
8.20
8.3
8.25
9.0
8.26
6.6
5.71
6.6
7.51
6.8
7.49
6.3–6.6
1.79
6.7
1.87
6.5
1.97
6.5
3J(H,H) Hz
m
7.85
m
8.01
m
9
6.78
8.5
6.85
8.4
7.24
8.1, 8.5
7.31
6.82
7.7
6.77
7.7
2.32
2.73
3J(H,H) Hz
1
3J(H,H) Hz
7.0, 8.8
7.3, 8.4
aAssignment in accordance to gHSQC, gHMBC and H,H-COSY experiments (m=multiplet).
Formation of saphenic acid derived Mosher’s esters was
accomplished in excellent yields but separation of the
diastereomeric esters failed by normal-phase as well as
reversed-phase chromatography.
intermediates and products were analyzed by chiral
HPLC and polarimetry. Table 2 shows a summary of
these results.
As can be seen from Table 2, a significant specific rota-
tion of more than 150 in chloroform was determined for
both enantiomers of saphenamycin. Earlier reports on
optical rotation of saphenamycin samples isolated from
natural sources range from 0 to +5 in chloroform.5,6,8
Partial to complete racemization during isolation, or the
existence of racemic saphenamycin in Nature could
explain this phenomena. The former is unlikely, since no
racemization was detected under reaction conditions
ranging from acidic (4 M HCl) to basic (pyridine,
DMAP, Et2NH, aq NaHCO3). The latter is question-
able as well, since the biosynthesis of saphenamycin is
believed to proceed via shikimic acid to the prochiral
Finally, our attention was drawn to resolution using
(À)-brucine, which had received substantial attention in
literature.22,23 Resolution of saphenic acid was com-
pleted by (À)-brucine salt formation and crystallization
from methanol. The free acids were liberated by treat-
ment with NH3 (25% aq solution) and isolated by
acidification with HCl followed by concentration. After
the first crystallization, enantiomeric purity of the acid
of upto 80% was obtained in high yield. Successive
crystallizations only improved the enantiomeric purity
slightly, but also lowered the recovered yield of both
enantiomers considerably. Attempts to directly resolve
saphenamycin by brucine salt formation were unsuc-
cessful.
The brucine salt precipitated from the resolution was
recrystallized from water/methanol and plate crystals
were obtained and analyzed by X-ray crystallography
(Fig. 2).
The absolute configuration of the precipitated (À)-bru-
cinium (À)-saphenate was determined to be the R-form
and (À)-saphenic acid had a specific rotation of À19.7
(c0.13) and +42.0 (c0.23) in chloroform and DMSO,
respectively. Geiger et al. reported a specific rotation of
saphenic acid isolated from a Streptomyces strain to be
À11.4 (c0.57) and +55 (c0.16) in chloroform and
DMSO, respectively.8 The small rotations can be
tedious to measure in dilute samples and increased con-
centration leads to high absorption at the wavelength of
the sodium-D line. This can explain the differences in
rotation determined and it can be concluded that
saphenic acid isolated from Streptomyces is the (R)-
enantiomer.
Synthesis of saphenamycin enantiomers
Figure 2. X-ray crystal structure of (À)-brucinium (À)-(R)-saphenate.
Ellipsoides are 50% probability levels.24 Hydrogen bonds between
carboxylate of saphenic acid and ammonium salt of brucine are
marked with thin lines and are 1.9 and 2.3 A, respectively. Refine-
ments were performed in SHELXL98 and the picture is created by
ORTEP II.
The resolved enantiomers of saphenic acid were carried
through the synthesis identical to the racemate (Scheme
2) and similar yields were obtained of the enantiomers
of saphenamycin. Enantiopurity was retained through-
out the synthesis and no racemization was observed. All

Jane B. Laursen et al. / Bioorg. Med. Chem. 11 (2003) 723–731
727
Table 2. Polarimetric results, melting points and retention times of enantiomers during saphenamycin synthesisa
[a]2D2
Melting point (ꢀC), (R/S)
Retention time (min.)c
b
Compound
Conformation
3
R
S
R
S
R
S
R
S
À19.7
+20.0
À27.3
204–206, (83:17)
203–205, (10:90)
99–101, (69:31)
102–104, (21:79)
Oil, (81:19)
Oil, (16:84)
159–164, (90:10)
162–165, (19:81)
28.8
33.0
11.2
12.7
13.2
14.9
54.2
49.2
4
9
1
+34.3
À53.5
+50.0
À150.6
+176.9
aSpecific rotations of enantiomeric pairs are not numerically identical. This is ascribed to the small amounts weighed, since high concentrations
resulted in poor transmittance at 589 nm.
bMeasured in chloroform using the sodium-D line (589 nm). Corrected according to enantiopurity (see Experimental).
cHPLC on Chiralcel OD-H column eluting with hexane/2-propanol/TFA (70:30:0.1).
ketone 6-aceto-phenazine-1-carboxylic acid.20 A sub-
sequent enzymatic reduction, which is presumably
enantioselective, leads to formation of saphenic acid.
Successive esterification to form saphenamycin is not
likely to deteriorate optical activity. Simple esterifica-
tion of the free carboxylic acid in saphenamycin isolated
from natural sources yields esters with significant spe-
cific rotation (>200 in chloroform).8 These arguments
indicate that saphenamycin in Nature is chiral and is
not likely to racemize under extraction and isolation
conditions. Hence, earlier reports on the specific rota-
tion appear questionable, and our measurements cannot
reveal the absolute configuration of the naturally
occurring species. Saphenic acid has now been identified
as the R-form, and it is therefore plausible that saphe-
namycin in Nature is also the R-form.
coccus aureus strains; clinical isolates of fusidic acid
resistant and rifampicin resistant MRSA (Table 3). The
zones of growth inhibition were sharpand clear indi-
cating bactericidal and not bacteristatic character of the
antibiotics. No significant difference was detected
between activities of enantiomers and racemate of saphe-
namycin against any organism, indicating that the stereo-
chemistry had no influence on activity. In all cases, the
antibiotic activity of saphenamycin was stronger or
equivalent to that of ceftriaxone and ciprofloxacin. MIC
values ranged from <0.1 to 0.3 mg/mL and were compar-
able to the values reported for natural isolates.5,6
Conclusions
The first reported synthesis of the natural antibiotic
saphenamycin was carried out with success starting
from synthetic saphenic acid and 6-methylsalicylic acid.
The racemate was synthesized with satisfactory yields
using a temporary allyl protection of carboxy and
phenoxy functionalities from racemic saphenic acid.
Biological activity
The importance of chirality for antimicrobial activity of
saphenamycin can be crucial information in determina-
tion of mode-of-action of this class of antibiotics. Seve-
ral reports on biological activities of saphenamycin
samples isolated from natural sources have appeared.5À8
These can be directly compared with our observations
of synthetic racemate as well as enantiomers.
Racemic saphenic acid was successfully resolved by
crystallization of the corresponding (À)-brucine dia-
stereomeric salts and the absolute configuration of
(À)-brucinium (À)-saphenate was determined by X-ray
crystallography to have R-configuration. Based on
earlier reported polarimetry analysis, we suggest that
Growth inhibition studies were performed on a panel of
Gram-positive skin flora and on two resistant Staphylo-
Table 3. Antibacterial activity of Saphenamycin racemate and enantiomersa
Organism/strain
MIC (mg/mL)
Natural 1
(Æ)-1
(À)-(R)-1c
(+)-(S)-1c
Referenced
Micrococcus luteus ATCC 9341
0.01–0.39b,6
0.015–2b,5
n.a.
0.01
0.01
0.03
0.19
0.05
0.11
0.20
0.09
0.11
0.01
0.02
0.04
0.30
0.16
0.17
0.16
0.16
0.11
0.04
0.02
0.06
0.30
0.12
0.17
0.22
0.26
0.18
<0.01
0.64
<0.001
1.84
0.76
0.10
0.18
0.24
0.32
Propionibacterium acnes ATCC 6919
Streptococcus pyogenes NCTC 8304
Streptococcus pyogenes ATCC 10403
Staphylococcus epidermidis ATCC 12228
Staphylococcus aureus ATCC 6538P, FDA 209P
Staphylococcus aureus ATCC 25923
Staphylococcus aureus CJ 234 (F) (MRSA)
Staphylococcus aureus CJ 234 (R) (MRSA)
n.a.
n.a.
0.396
n.a.
n.a.
n.a.
aMRSA, meticilline resistant S. aureus; F, fusidic acid resistant clinical isolate; R, rifampicin resistant clinical isolate; n.a., not available.
bMIC reported in literature on similar genera but not identical strains.
cEnantiopurity: (À)-(R)-1 (81% R, 19% S), (+)-(S)-1 (21% R, 79% S).
dCeftriaxone (susceptible strains), ciprofloxacin (resistant strains).

728
Jane B. Laursen et al. / Bioorg. Med. Chem. 11 (2003) 723–731
this enantiomer is the naturally occurring stereoisomer
of saphenic acid. The enantiomers of saphenamycin
were synthesized from the corresponding saphenic acid
enantiomers, also with satisfactory yields.
254 nm by Varian 9065 Polychrom UV detector. 1H and
13C NMR spectra were recorded on a Bruker AC200, a
Varian Mercury 300 or a Varian Unity Inova 500 spec-
trometer. The chemical shifts are referred to an internal
standard of CDCl3. gHMBC and H,H-COSY spectro-
scopy were recorded on a Varian Mercury 300 appara-
tus. Mass spectrometry was performed on a VG
Masslab trio-2 apparatus (EI–MS) and accurate mass
determination (high resolution MS, HR-MS) was per-
formed on a Micromass LCT apparatus with an AP-
ESI probe.
Optical studies of earlier isolated samples of saphena-
mycin have been inconsistent. Based on the knowledge
of the biosynthetic pathway and on determination of
significant specific rotations of the synthetic enantio-
mers, it is believed that saphenamycin in Nature exists
as one enantiomer. Thus, we suggest that the natural
configuration of this enantiomer is equivalent to its
parent saphenic acid, namely the R-form.
The enantiomers of saphenamycin were screened
against a range of normal skin flora and resistant S.
aureus strains. These were compared to the synthetic
racemate as well as earlier reported biological activities
of saphenamycin isolated from natural sources. No sig-
nificant difference was observed in the antibacterial
activity of the enantiomers of saphenamycin, thus it can
be concluded that the chirality of saphenamycin is of
less importance to its biological activity. Saphenamycin
proved to be a potent antibiotic against fusidic acid and
rifampicin resistant S. aureus strains showing MIC
values between 0.1 and 0.2 mg/mL.
2-Allyloxy-6-methylbenzoic acid ethyl ester (7). A mix-
ture of 2-hydroxy-6-methylbenzoic acid ethyl ester (6)
(9.0 g, 50 mmol, 1 equiv), K2CO3 (8.3 g, 60 mmol, 1.2
equiv), allyl bromide (8.5 mL, 100 mmol, 2 equiv) and
acetone (70 mL) was refluxed for 24 h. The mixture was
concentrated in vacuo, diluted with EtOAc, washed
with H2O and brine, dried with Na2SO4 and con-
centrated to give the title compound 7 as a yellow oil
1
(10.3 g, 94%). H NMR (300 MHz, CDCl3) d 1.37 (t,
J=7.2 Hz, CH3CH2, 3H), 2.30 (s, H-700, 3H), 4.39 (q,
J=7.2 Hz, CH2CH3, 2H), 4.55 (dt, J=1.4–1.9, 5.0 Hz,
OCH2CH¼CH2, 2H), 5.24 (dd, Jgem=1.4, Jcis=10.6
Hz, CH¼CH2, 1H), 5.38 (dd, Jgem=1.7 Hz, Jtrans=17.2,
CH¼CH2, 1H), 5.95-6.05 (m, 1H), 6.74 (d, J=8.5 Hz,
H-300, 1H), 6.79 (d, J=7.5 Hz, H-500, 1H), 7.19 (t, J=8.0
Hz, H-3‘, 1H) ppm. 13C NMR (75 MHz, CDCl3) d 14.6
(CH3CH2), 19.4 (C-700), 61.2 (CH2CH3), 69.5
(OCH2CH¼CH2), 110.2 (C-300), 117.3 (CH2¼CH),
122.8 (C-500), 124.8 (C-100), 130.2 (C-400), 133.2
(CH¼CH2), 136.6 (C-600), 155.6 (C-200), 168.4 (C¼O)
ppm. EI–MS calcd m/z: 220, found 220 [M]+.
Further studies in our laboratories will elucidate the
mode-of-action of this class of molecules. Different
structural variations in order to improve activity or even
enhance specificity will be investigated.
Experimental
Commercially available reagents were used without
further purification unless otherwise noted. Saphenic
acid14 and 6-methylsalicylic acid ethyl ester13 were
prepared as previously reported. Solvents for reac-
tions were distilled prior to use and stored over
molecular sieves. Flash chromatography was per-
formed manually on silica gel (Matrex 60 A, 35–70
m) or automatically on a Biotage Quad3+ apparatus
with prepacked columns (KP-SIL, 32–63 mm, 60 A).
Thin layer chromatography was performed on Merck
silica gel 60 F254 alumina plates. Corrected melting
points were measured in open capillary tubes. Analy-
tical HPLC was performed using Merck Hitachi
LaChrom system (pump L-7100, autosampler L-7200,
column oven L-7300, diode array detector L-7450) with
a Supercosil RP-3,3 column. Gradients of 0.035% TFA
in CH3CN (A) and 0.05% aqueous TFA (B) with a flow
of 1.5 mL/min were used as follows: 100% B for 30 s;
from 100% B to 100% A over 10 min; 100% A for 2
min; from 100% A to 100% B over 6 s; 100% B for 2
min. Preparative HPLC was performed on a Waters 600
system with Waters 996 diode array detector and three
consecutive columns (40Â100 mm prep. NOVA Pak
HR C18 6 mm 60 A units). Linear gradients of CH3CN
and water (MilliQ) were used. Chiral HPLC was per-
formed on a Chiralcel OD-H column (0.46 cm ØÂ25
cm) eluting with hexane/2-propanol/TFA (70:30:0.1) in
a Varian 9012 Solvent Delivery System and visualized at
2-Allyloxy-6-methylbenzoic acid (8). A homogeneous
mixture of 2-allyloxy-6-methylbenzoic acid ethyl ester
(7) (7.7 g, 35 mmol, 1 equiv), NaOH (2.8 g, 70 mmol, 2
equiv), H2O (20 mL) and EtOH (20 mL) was refluxed
for 30 h. After 2 h at reflux the mixture shifted from
yellow to a brown turbid color and after a total of 30 h
at reflux it had turned into a clear yellow color. The
mixture was concentrated, acidified using concentrated
HCl (aq) and extracted with EtOAc (3Â25 mL). The
combined organic phases were concentrated in vacuo
1
yielding the product 8 as a yellow oil (6.6 g, 98%). H
NMR (300 MHz, CDCl3) d 2.48 (s, H-700, 3H), 4.64 (dt,
J=1.5, 5.1 Hz, OCH2CH¼CH2, 2H), 5.31 (dd,
Jgem=1.2–1.5, Jcis=10.4 Hz, CH¼CH2, 1H), 5.44 (dd,
Jgem=1.3–1.8 Hz, Jtrans=17.2, CH¼CH2, 1H), 5.98–
6.09 (m, 1H), 6.82 (d, J=8.4 Hz, H-300, 1H), 6.88 (d,
J=7.5 Hz, H-500, 1H), 7.28 (dd, J=7.5, 8.4 Hz, H-400,
1H) ppm. 13C NMR (50 MHz, CDCl3) d 20.2 (C-700),
69.7 (OCH2CH¼CH2), 110.2 (C-300), 117.9 (CH2¼CH),
121.9 (C-100), 123.5 (C-500), 131.0 (C-400), 132.4
(CH¼CH2), 138.6 (C-600), 155.9 (C-200), 171.7 (C¼O)
ppm. EI–MS calcd m/z: 192, found 192 [M]+.
Racemic 6-(1-hydroxy-ethyl)-phenazine-1-carboxylic acid
allyl ester, allyl saphenate ((Æ)-4). To a flask containing
racemic saphenic acid (Æ)-3 (400 mg, 1.5 mmol, 1
equiv) under argon was added allyl bromide (3 mL, 35.5

Jane B. Laursen et al. / Bioorg. Med. Chem. 11 (2003) 723–731
729
mmol, 24 equiv) and CH3CN (2Â1 mL). DIEA (255 mL,
1.5 mmol, 1 equiv) was added dropwise. The mixture
was left stirring at ambient temperature under argon for
18 h. The mixture was diluted with EtOAc, washed with
10% Na2CO3 (aq, 3Â) and brine, dried with MgSO4
and concentrated in vacuo. Flash chromatography (3:2
hexane/EtOAc) gave the product, which was recrys-
tallized from EtOAc/hexane to give (Æ)-4 as fine yellow
crystals in quantitative yield (460 mg). Mp87–88 ꢀC.
Rf=0.3 (2:1 hexane/EtOAc). Anal. HPLC (251 nm,
>99%, 220 nm, >95%) RT=5.71 min. 1H NMR
(300 MHz, CDCl3) d 1.79 (d, J=6.7 Hz, H-20, 3H), 4.83
(br. s, OH, 1H), 5.04 (dt, J=1.5, 5.6 Hz, CH2CH¼CH2,
2H), 5.36 (dq, Jcis=1.4, 10.4 Hz, CH¼CH2, 1H), 5.60
(dq, Jtrans=1.7, 17.2 Hz, CH¼CH2, 1H), 5.71 (q, J=6.6
Hz, H-10, 1H), 6.09–6.21 (m, 6.15, CH¼CH2, 1H), 7.77–
7.88 (m, H-3, H-7, H-8, 3H), 8.20 (d, J=8.3 Hz, H-9,
1H), 8.26 (dd, J=1.2, 6.9 Hz, H-2, 1H), 8.34 (dd, J=1.4,
8.7 Hz, H-4, 1H) ppm. 13C NMR (75 MHz, CDCl3) d 23.9
(C-20), 66.4 (OCH2CH¼CH2), 68.9 (C-10), 118.7
(OCH2CH¼CH2), 127.4 (C-7), 129.5 (C-3), 129.7 (C-9),
131.0 (C-8), 131.6 (C-1), 132.3 (OCH2CH¼CH2), 132.4
(C-2), 133.5 (C-4), 140.8 (C-6), 141.0 (C-10a), 142.1 (C-
5a), 142.9 (C-4a), 144.3 (C-9a), 166.4 (C¼O). HR-MS
calcd m/z: 309.1239, found 309.1036 [M+H]+. Elemental
analysis for C18H16N2O3: C, 70.12; H 5.23; N, 9.09.
Found: C, 69.96; H, 5.19; N, 9.04.
110.7, 115.7, 118.4, 120.5, 122.9, 127.0, 127.5, 128.7,
128.8, 130.1, 130.4, 130.7, 130.8, 131.4, 132.0, 133.9,
135.0, 138.6, 139.5, 140.2, 140.8, 141.3, 141.9, 143.1,
147.4, 163.3, 165.7, 166.4, 170.3 ppm. HR-MS calcd
m/z: 577.1975, found 577.2192 [M+H]+.
Only small amounts (18.1 mg, 32%) of the allyl ester of
saphenamycin were isolated.
Racemic
6-[1-(2-allyloxy-6-methyl-benzoyloxy)-ethyl]-
phenazine-1-carboxylic acid allyl ester ((Æ)-9). To a
mixture of 2-allyloxy-6-methylbenzoic acid (8) (250 mg,
1.301 mmol, 1 equiv), DMF (three drops), and CH2Cl2
(1 mL) was added oxalyl chloride (120 mL, 1.418 mmol,
1.1 equiv) dropwise with a syringe under inert condi-
tions. The mixture was left stirring for 2 h at ambient
temperature and the solvent was evaporated under inert
conditions. A solution of racemic allyl saphenate (Æ)-4
(200 mg, 0.649 mmol, 0.5 equiv) and DMAP (8 mg,
0.065 mmol, 0.05 equiv) in pyridine (2.5 mL) was added
to the 2-allyloxy-6-methylbenzoic acid chloride, and
evolution of HCl (g) was observed. The mixture was left
stirring over night under argon at ambient temperature.
The color changed from yellow to dark brown. The mix-
ture was held over ice water and acidified until pH ꢁ5.
The aqueous phase was extracted with EtOAc (3Â). The
organic phases were combined, dried over MgSO4, con-
centrated in vacuo and purified by preparative HPLC
yielding the product (Æ)-9 as a yellow oil (200 mg, 64%).
28% (Æ)-4 was recovered. Anal. HPLC (251 nm, >99%,
220 nm, >97%) RT=8.01 min. 1H NMR (500MHz,
CDCl3) d 1.87 (d, J=6.5 Hz, H-20, 3H), 2.32 (s, H-700, 3H),
4.56 (ddd, J=1.7, 5.1 Hz, OCH2CH¼CH2, 2H), 5.05
(ddd, J=1.3–1.7, 5.5 Hz, C(O)OCH2CH¼CH2, 2H),
Racemic
6-{1-[2-(2-hydroxy-6-methyl-benzoyloxy)-6-
methyl-benzoyloxy]-ethyl}-phenazine-1-carboxylic acid
allyl ester ((Æ)-5). To a mixture of 6-methylsalicylic
acid (2) (79 mg, 0.519 mmol, 1 equiv), DMF (two
drops), and CH2Cl2 (0.5 mL) was added oxalyl chloride
(50 mL, 0.571 mmol, 1.1 equiv) dropwise with a syringe
at 0 ꢀC under inert conditions. The mixture was left
stirring for 2 h and allowed to reach room temperature.
A solution of racemic allyl saphenate (Æ)-4 (40 mg,
0.130 mmol, 0.25 equiv) and DMAP (19.8 mg, 0.162
mmol, 0.31 equiv) in CH2Cl2 (0.5 mL) was added to the
6-methylbenzoic acid chloride, and evolution of HCl (g)
was observed. The mixture was left stirring over night
under argon at ambient temperature, diluted with
CH2Cl2, washed with 2 N HCl (aq), H2O and brine,
dried with MgSO4, concentrated in vacuo and purified
by automated column chromatography eluting with
100% CH2Cl2. The major product was the ‘double
ester’ 5 (48.1 mg, 64%). Anal. HPLC (251 nm, >93%)
5.20
(dq,
Jgem=1.3–1.7
Hz,
Jcis=10.7
Hz,
OCH2CH¼CH2, 1H), 5.35 (dq, Jgem=1.7 Hz,
Jtrans=17.1 Hz, OCH2CH¼CH2, 1H), 5.37 (dq, Jgem=1.3
Hz, Jtrans=10.7 Hz, C(O)OCH2CH¼CH2, 1H), 5.62 (dq,
Jgem=1.3–1.7, Jtrans=17.5 Hz, C(O)OCH2CH¼CH2,
1H), 5.92–6.00 (m, OCH2CH¼CH2, 1H), 6.12–6.20 (m,
C(O)OCH2CH¼CH2, 1H), 6.78 (d, J=8.5 Hz, H-300, 1H),
6.82 (d, J=7.7 Hz, H-500, 1H), 7.24 (dd, J=8.1, 8.5 Hz, H-
400, 1H), 7.51 (q, J=6.8 Hz, 1H), 7.85 (m, H-3 and H-8,
2H), 8.04 (d, J=6.5 Hz, H-7, 1H), 8.25 (d, J=9.0 Hz, H-9,
1H), 8.28 (d, J=6.8 Hz, H-2, 1H), 8.46 (d, J=9.0 Hz, H-4,
1H) ppm. 13C NMR (125 MHz, CDCl3) d 19.9 (C-700),
22.9 (C-20), 66.8 (C(O)OCH2CH¼CH2), 69.7
RT=8.27 min. H NMR (500 MHz, CDCl3) d 1.71 (d,
(OCH2CH¼CH2),
70.1
(C-10),
110.6,
118.3
1
J=6.7 Hz, H-20, 3H), 2.33 (s, H-7000, 3H), 2.49 (s, H-700,
3H), 5.05 (dt, J=1.2–1.8, 5.5 Hz, OCH2CH¼CH2, 2H),
5.39 (dq, Jgem=1.2–1.3 Hz, Jcis=10.3 Hz, 1H), 5.63 (dq,
Jgem=1.3–2 Hz, Jtrans=16.8 Hz, C(O)OCH2CH¼CH2,
1H), 6.13–6.21, (m (ddt centered at 6.17 ppm),
OCH2CH¼CH2, 1H), 6.24 (d, J=7.7 Hz, H-3000, 1H),
6.65 (d, J=8.4 Hz, H-5000, 1H), 7.02 (d, J=8.1 Hz, H-300
or H-500, 1H), 7.07 (t, J=7.8–7.9 Hz, H-4000, 1H), 7.23 (d,
J=7.8 Hz, H-300 or H-500, 1H), 7.42 (t, J=7.9 Hz, H-400,
1H), 7.52 (q, J=6.1–6.6 Hz, H-10, 1H), 7.68 (dd, J=7.0,
8.7 Hz, H-8, 1H), 7.81 (d, J=6.8 Hz, H-7, 1H), 7.85
(dd, J=7.0, 8.8 Hz, H-3, 1H), 8.09 (dd, J=1.4, 8.4 Hz,
H-9, 1H), 8.26 (dd, J=1.6, 7.1 Hz, H-2, 1H), 8.39 (dd,
J=1.4, 9.1 Hz, H-4, 1H), 10.87 (s, OH, 1H) ppm. 13C
NMR (125 MHz, CDCl3) d 20.0, 21.7, 24.0, 66.1, 68.9,
(OCH2CH¼CH2), 119.1 (C(O)OCH2CH¼CH2), 123.3,
125.2, 127.6, 129.5, 130.5, 130.8, 131.4, 132.0, 132.8,
132.9, 133.6, 134.7, 137.3, 141.4, 141.5, 141.6, 142.7,
144.2, 156.3 (arom. C–O), 167.1 (C(O)O), 168.1
(C(O)OCH2CH¼CH2) ppm. HR-MS calcd m/z:
483.1920, found 483.1951 [M+H]+.
Racemic
6-[1-(2-hydroxy-6-methyl-benzoyloxy)-ethyl]-
phenazine-1-carboxylic acid ((Æ)-1). Compound (Æ)-9
(49 mg, 102 mmol, 1 equiv) and Pd(PPh3)4 (17.6 mg, 15.3
mmol, 0.15 equiv) were dissolved in dioxane (1 mL)
under argon, a red color evolved. Et2NH (24 mL, 230
mmol, 2.25 equiv) was added as an external nucleophile.
After 3 h the reaction had ceased and the color changed
into brown-green. The reaction mixture was diluted with

730
Jane B. Laursen et al. / Bioorg. Med. Chem. 11 (2003) 723–731
CH2Cl2 and washed with 2 N HCl (aq) (3Â), H2O (3Â)
and brine. The organic phase was dried with MgSO4, fil-
tered, concentrated in vacuo and purified by automated
column chromatography eluting with 100% CH2Cl2. 9
and the allyl aryl intermediate 6-[1-(2-allyloxy-6-methyl-
benzoyloxy)-ethyl]-phenazine-1-carboxylic acid (9a) were
both at Rf=0.21 (100% CH2Cl2), but separation was
visualized on TLC using CH2Cl2/diethyl ether (2:1) [Rf
(9)=0.56; Rf (9a)=0.63]. Racemic saphenamycin (Æ)-1
was obtained in 61% yield (25 mg) as yellow-green
crystals (using non-optimized conditions). Mp 178–
180 ꢀC. Anal. HPLC (251 nm, >90%, 220 nm, >84%)
31% S) RT (R)=11.2 min. Mp99–101 ꢀC. [a]D22 À27.3
(c0.79, CHCl3).
(+)-(S)-4 was obtained from (+)-(S)-3 in 82% yield as
yellow crystals. Anal. chiral HPLC (254 nm: 21% R,
79% S) RT (S)=12.7 min. Mp102–104 ꢀC. [a]D22+34.3
(c0.70, CHCl3).
(À)-(R)- and (+)-(S)-6-[1-(2-allyloxy-6-methyl-benzoy-
loxy)-ethyl]-phenazine-1-carboxylic acid allyl ester (9).
Procedure: see racemic (Æ)-9.
1
RT=7.37 min. H NMR (500 MHz, CDCl3) d 1.97 (d,
(À)-(R)-9 was obtained from (À)-(R)-4 and 7 in 40%
yield as a yellow oil. 16% (À)-(R)-4 was recovered.
Anal. chiral HPLC (254 nm: 81% R, 19% S) RT
(R)=13.2 min. [a]2D2 À53.5 (c0.65, CHCl3).
J=6.5 Hz, H-20, 3H), 2.73 (s, H-700, 3H), 6.77 (d, J=7.7
Hz, H-500, 1H), 6.85 (d, J=8.4 Hz, H-300, 1H), 7.31 (t,
J=7.3, 8.4 Hz, H-400, 1H), 7.49 (q, J=6.3-6.6 Hz, H-100,
1H), 8.00–8.01 (m, H-7 and H-8, 2H), 8.08 (dd, J=7.0,
8.8 Hz, H-3, 1H), 8.26 (dd, J=3.3, 6.6 Hz, H-9, 1H),
8.61 (dd, J=1.5, 8.8 Hz, H-4, 1H), 9.02 (dd, J=1.5, 7.0
Hz, H-2, 1H), 11.12 (s, OH, 1H), 15.44 (br. s, COOH,
1H) ppm. 13C NMR (125 MHz, CDCl3) d 22.4, 24.7,
70.3, 112.5, 116.1, 123.3, 125.1, 127.8, 128.0, 130.6,
133.1, 134.6, 135.6, 137.9, 140.0, 140.2, 141.2, 141.4,
141.5, 142.8, 163.2, 166.0, 170.9 ppm. HR-MS calcd m/
z: 403.1294, found 403.1335 [M+H]+.
(+)-(S)-9 was obtained from (+)-(S)-4 and 7 in 55%
yield as a yellow oil. 31% (+)-(S)-4 was recovered.
Anal. chiral HPLC (254 nm: 16% R, 84% S) RT
(S)=14.9 min. [a]2D2+50.0 (c0.63, CHCl3).
Synthesis of (À)-(R)- and (+)-(S)-6-[1-(2-hydroxy-6-
methyl-benzoyloxy)-ethyl]-phenazine-1-carboxylic acid
(1). Procedure: see racemic (Æ)-1.
Resolution of (Æ)-saphenic acid, 6-(1-hydroxyethyl)-
phenazine-1-carboxylic acid ((Æ)-3). The racemic acid
(Æ)-3 (537 mg, 2 mmol, 1 equiv) was dissolved in mini-
mum amount hot MeOH (25 mL). (À)-Brucine (789 mg,
2 mmol, 1 equiv) was added and dissolved immediately.
The mixture was allowed to cool to room temperature
and left at 4ꢀC for 30 h in a sealed flask containing
water in order to allow diffusion between the solvents.
The mixture was filtered and the mother liquor was
concentrated in vacuo. Both portions of crystals were
stirred in NH4OH (25% solution, 50 mL) for 1 h at
ambient temperature. The free brucine was removed by
extraction with CH2Cl2 (Â3). The aqueous phase was
acidified by 4 M HCl (aq) to pH 4, extracted with
CH2Cl2 (Â3), dried over MgSO4 and concentrated in
vacuo to yield the free acid.
(À)-(R)-1 was obtained from (À)-(R)-9 in 88% yield as
yellow-green crystals. Mp. 159–164 ꢀC. Anal. chiral
HPLC (254 nm: 90% R, 10% S) RT(R)=54.2 min. [a]D22
À150.6 (c=0.79, CHCl3).
(+)-(S)-1 was obtained from (+)-(S)-9 in 71% yield as
yellow-green crystals. Mp162–165 ꢀC. Anal. chiral
HPLC (254 nm: 22% R, 78% S) RT (S)=49.2 min.
[a]2D2+176.9 (c0.34, CHCl3).
Antimicrobial assay. Minimum inhibitory concentrations
(MICs) were estimated using an agar cupassay. Bacterial
strains were obtained form the American Type Culture
Collection (ATCC) or from the collection of clinical iso-
lates at Leo Pharma A/S. Colonies from fresh overnight
cultures were resuspended in saline water to 0.5 MacFar-
land corresponding to 108 CFU/mL. A total of 200 mL
Mueller Hinton agar (Oxoid) at 48 ꢀC was inoculated at a
concentration of 106 CFU/mL and poured into square
Petri dishes (245Â245 mm). Holes were made in the
inoculated plates and 200 mL of each sample was disposed
into each hole. Dilution series for Saphenamycin (race-
mate and enantiomers) contained six dilutions between
0.25 and 125 mg/mL. For Streptococci, Mueller Hinton
agar was supplemented with 5% sheep blood. Plates were
appropriately incubated and zone diameters of growth
inhibition were measured using an electronic caliper.
MIC values were estimated using linear regression curve
between the zone diameter of growth inhibition and the
log2 of the sample concentration. MIC values lower than
0.25 mg/mL were determined by extrapolation.
(À)-(R)-3 was obtained in 92% yield as yellow crystals
from the less soluble salt. Anal. chiral HPLC (254 nm:
ꢀ
83% R, 17% S) RT (R)=28.8 min. Mp204–206 C. [a]D22
À19.7 (c0.13, CHCl3), [a]2D2+42.0 (c0.23, DMSO). Cal-
culations of these specific rotations as well as the fol-
lowing are based on enantiopurity of the sample
measured, thus resemble the specific rotation of the pure
enantiomer.
(+)-(S)-3 was obtained in 70% yield as yellow crystals
from the more soluble salt. Anal. chiral HPLC (254 nm:
10% R, 90% S) RT (S)=33.0 min. Mp. 203–205 ꢀC.
[a]2D2+20.0 (c0.41, CHCl3).
(À)-(R)- and (+)-(S)-allyl saphenate, (À)-(R)- and (+)-
(S)-6-(1-hydroxy-ethyl)-phenazine-1-carboxylic acid allyl
ester (4). Procedure: see racemic (Æ)-4.
Acknowledgements
(À)-(R)-4 was obtained from (À)-(R)-3 in 84% yield as
yellow crystals. Anal. chiral HPLC (254 nm: 69% R,
Professor Knud J. Jensen (The Royal Veterinary and
Agricultural University) is acknowledged for his effort

Jane B. Laursen et al. / Bioorg. Med. Chem. 11 (2003) 723–731
731
in obtaining and discussing assignment of 500 MHz
NMR spectra. Professor Pher G. Andersson is
acknowledged for kindly providing us with laboratory
facilities in Uppsala University for the initial studies of
chiral resolution by semi-preparative HPLC. Professor
Sine Larsen’s groupat University of Coepnhagen,
especially Mr. Flemming Hansen and Ms. Jette Odder-
shede, are acknowledged for their contributions to
structure elucidation by X-ray. Professor Per-Ola
Norrby (Technical University of Denmark) is acknowl-
edged for fruitful discussions regarding Pd-mediated
deprotection. Dr. Hanne Theilgaard (Leo Pharma A/S)
is acknowledged for performing biological assays.
Finally, we thank the Technical University of Denmark
(JBL), European Commission COST D16’s Short-Term
Scientific Mission Program (JBL, Uppsala), Novo Nor-
disk Training and Research Programme (CGJ) and the
Villum Kann Rasmussen foundation (JN) for financial
support.
8. Geiger, A.; Keller-Schierlein, W.; Brandl, M.; Zahner, H. J.
Antibiot. 1988, 41, 1542.
9. Kerr, J. R. Infect. Dis. Rev. 2000, 2, 184.
10. Laursen, J. B.; de Visser, P.; Nielsen, H. K.; Jensen, K. J.;
Nielsen, J. Bioorg. Med. Chem. Lett. 2002, 12, 171.
11. Hollstein, U.; Van Gemert, R., Jr. J. Biochem. 1971, 10,
497.
12. Mosher, C. W.; Lee, D.-Y.; Enanoza, R. M.; Sturm, P. A.;
Kuhlmann, K. F. J. Med. Chem. 1979, 22, 918.
13. Hauser, F. M.; Pogany, S. A. Synthesis 1980, 814.
14. Petersen, L.; Jensen, K. J.; Nielsen, J. Synthesis 1999, 10,
1763.
15. Bahnmuller, U.; Keller-Schierlein, W.; Brandl, M.; Zah-
ner, H.; Diddens, H. J. Antibiot. 1988, 41, 1552.
16. Wissner, A.; Grudzinskas, C. V. J. Org. Chem. 1978, 43,
3972.
17. Greene, T. W.; Wuts, P. G. M. Protective Groups in
Organic Synthesis, 3rd ed.; John Wiley & Sons: New York,
1999.
18. Alsina, J.; Rabanal, F.; Chiva, C.; Giralt, E.; Albericio, F.
Tetrahedron 1998, 54, 10125.
19. All other NMR data were similar to the corresponding
acid. For related example, see: Islam, R., Joule, J. A. J. Chem.
Res. (M) 1990, 364.
20. Van’t Land, C. W.; Mocek, U.; Floss, H. G. J. Org.
Chem. 1993, 58, 6576.
References and Notes
1. Helene, C. Nature 1998, 391, 436.
2. Cao, Y. and He, X.-W. Spectrochim. Acta, Part A 1998, 54,
883
3. Berman, H. M.; Young, P. R. Ann. Rev. Biophys. Bioeng.
1981, 10, 87.
21. Yun, B.-S.; Ryoo, I.-J.; Kim, W.-G.; Kim, J.-P.; Koshino,
H.; Seto, H.; Yoo, I.-D. Tetrahedron Lett. 1996, 37, 8529.
22. Ezquerra, J.; Yyruretagoyena, B.; Avendano, C.; de la
Cuesta, E.; Gonzalez, R.; Prieto, L.; Pedregal, C.; Espada, M.;
Prowse, W. Tetrahedron 1995, 51, 3271.
4. Neidle, S.; Laughton, C. A. In The Search for New Anti-
cancer Drugs; Waring, M. J., Ponder, B. A. J., Eds.; Kluwer
Academic: Lancaster, UK, 1992; p133.
5. Michel, K. H.; Hoehn, M. M. US Patent 4,316,959, 1982;
Chem. Abstr., 1982, 96, 197888.
6. Kitahara, M.; Nakamura, H.; Matsuda, Y.; Hamada, M.;
Naganawa, H.; Maeda, K.; Umezawa, H.; Itaka, Y. J. Anti-
biot. 1982, 1412.
7. Keller-Schierlein, W.; Geiger, A.; Zahner, H.; Brandl, M.
Helv. Chim. Acta 1988, 71, 2058.
23. Tichy, M.; Zavada, J.; Podlaha, J.; Vojtısek, P. Tetra-
hedron: Asymmetry 1995, 6, 1279.
24. Crystallographic data (excluding structure factors) for the
structure in this paper has been deposited with the Cambridge
Crystallographic Data Centre as supplementary publication
nos. CCDC 187894. Copies of the data can be obtained, free
of charge, on application via www.ccdc.cam.ac.uk/conts/
retrieving.html (or from the CCDC, 12 Union Road, Cambridge
CB2 1EZ, UK, fax: +44-1223-336033, e-mail: deposit
@ccdc.cam.ac.uk).